LG 42LE4500 - No 20V rail

[DarkNeo]

I'm trying to fix an LG 42LE4500-ZA TV for a friend, and I was hoping someone might be able to help. At first glance it looked like the backlight had gone, but on further investigation it turns out the 20V power output from the PSU isn't working.

The PSU has the Model: YP42LPBA, Part Number: EAY60803202

The 3.5V output is working, so the MICOM on the main board powers up and will attempt to power on the TV when the power button is pressed. Power On will go high, the front panel LEDs will do their power on dance and the relay will click on, but there's 0V on the 20V lines.

I've attached several images including an annotated one which I'll refer to.

I've narrowed down the issue to the top section of the board, as there's a fused 50V input to this section which shows no ripple on the oscilloscope (green point - EL142).

I've tested the purple points, the 20V output pins on P201 and the inputs and output of rectifier D232, these show 0V with no noticable noise/ripple above the noise from my cheap oscilloscope.

The highlighted packages are:

Blue: STF13NM60N MOSFET https://cdn.badcaps-static.com/pdfs/...ba5c2ba3ce.pdf
Red: MBRF20100CTG Rectifier https://cdn.badcaps-static.com/pdfs/...e518a26d5c.pdf
Yellow: ST L6599 Resonant Controller https://cdn.badcaps-static.com/pdfs/...7cd0b455f4.pdf

So this is an LLC resonant bridge converter apparently, which I've done some research on and concluded that I have no idea how an LLC resonant bridge converter works or how to debug one!

I did try probing the inputs to the lower transformer and got the attached results (one in standby and the other 'on'). It doesn't look right to me, but I wouldn't know what does look right either.

I can replace the PSU as a last resort, but if it's a component I'd rather just replace that (unless it's one of the transformers, I can't find sources for them anywhere).

[Andrew F. Ali]

Re: LG 42LE4500 - No 20V rail

The 20Vdc would be on the COLD side of the PSU. Are you getting the 20Vac out from the transformer?

[DarkNeo]

Re: LG 42LE4500 - No 20V rail

I'm checking for 20Vdc on the output of the rectifier on the cold side (getting 0Vdc), and I'm looking for 20Vac on the inputs of the rectifier which are connected directly to the output of the transformer (getting 0Vac), but I've also measured from the pins of the transformer directly to be sure.

I'm wondering if I'm not doing the measurement correctly because I'm so used to DC. I'm connecting the reference wire of my probe to the chassis ground - which I've checked is connected to the GND pins on P201. But that's DC, do I need a different 0V reference for the AC?

Also I completely forgot to check the 12V rail, this is also showing 0Vdc on the output.

[Andrew F. Ali]

Re: LG 42LE4500 - No 20V rail

Are Q102 and Q103 getting power at the S pins? IC 101 could be malfunctioning, too.

[DarkNeo]

Re: LG 42LE4500 - No 20V rail

I attached the probes to both Q102 and Q103 S pins and the result looks very similar to the input to the transformer (see attached). Voltage doesn't go much above 0, but goes as low as -50V.

From what I gathered from IC101s datasheet Q102 and Q103 should be alternating, but they're in perfect step.

[budm]

Re: LG 42LE4500 - No 20V rail

1) It looks like you use chassis GND for your scope, the negative leg of the main filter cap is the GND ref for meter and scope, you better make sure that the TV is plugged into the isolation transformer when using the scope on the hot side otherwise you will see nice smoke and spark if the scope GND is tied to the scope incoming AC SAFETY GND.
2) "I've narrowed down the issue to the top section of the board, as there's a fused 50V input to this section which shows no ripple on the oscilloscope (green point - EL142)."
it shows 50V because you use the wrong GND as explained above, it should have about 390VDC with ref to the negative legs of the main filter caps (those lytics caps with 450V rating on them).
3) We also need the Voltage readings of all the pins of the connector P201.

[DarkNeo]

Re: LG 42LE4500 - No 20V rail

1) It looks like you use chassis GND for your scope, the negative leg of the main filter cap is the GND ref for meter and scope, you better make sure that the TV is plugged into the isolation transformer when using the scope on the hot side otherwise you will see nice smoke and spark if the scope GND is tied to the scope incoming AC SAFETY GND.
2) "I've narrowed down the issue to the top section of the board, as there's a fused 50V input to this section which shows no ripple on the oscilloscope (green point - EL142)."
it shows 50V because you use the wrong GND as explained above, it should have about 390VDC with ref to the negative legs of the main filter caps (those lytics caps with 450V rating on them).

Thanks, I clearly need to learn more on this so I'll do some research before trying anything with the scope. I take it if I'm using a battery powered multimeter I can take primary side readings between the point I'm checking and the negative main filter cap leg without any issues?

3) We also need the Voltage readings of all the pins of the connector P201.

All these readings are taken when the set is 'on':
#1 Power On: 3.41V
#2/3/4 20V: 0.02V
#5/6/7/8 GND: 0V
#9/10/11/12 3.5V: 3.51V
#13/14/15 GND: 0V
#16 V-sync: 0V
#17/19/21 12V: 0.1V
#18 DRV On: 0.01V
#20 SCLK: 0.03V
#22 PWM DIM: 0V
#23 SIN: 0.04V
#24 Error Out: 0V

[budm]

Re: LG 42LE4500 - No 20V rail

At this point do not worrying about using the scope yet, yes you can use battery operate on the hot side but you have to remember that the scope GND will now be at HOT potential so do not touch any the metal parts, scope probe with your hands or body.
I would verify that you do have 390VDC through that fuse to run the 12V/20V/ LED driver power supply or not first.
It also looks like some of the control signals are missing from the main board:
#18 DRV On (BL-ON): 0.01V
#22 PWM DIM: 0V
#20 SCLK: 0.03V don't know if it is used or not
#23 SIN: 0.04V, don't know what that is.
#24 Error Out: 0V, don't know if 0 = Normal or = Bad

Need to see if we can find service manual for it.

[DarkNeo]

Re: LG 42LE4500 - No 20V rail

Ok I'm reading 393VDC on that fuse.

I have the service manual for the 42LE4500 with schematics for the main board - unfortunately not the PSU.

#18 DRV On - goes to the INV_CTL pin on the MICOM
#22 PWM DIM - goes to the LVDS and PWM2 on the MStar CPU
#20 SCLK & #23 SIN - according to the schematic and option resistors on the board these pins are N/C
#24 Error Out - I'm not sure on this either, all I know is that it's an output from the LED driver daughterboard on the PSU

The only thing that I can see from the control signals that's not immediately explained by the lack of 12/20V rails to power the CPU is the DRV/BL On signal, but the MICOM may not activate that until told to do so by the CPU.

[budm]

Re: LG 42LE4500 - No 20V rail

Once the PS-ON is present the 12V and the 20V should be present, so we need to get the P/N of IC101 to get the spec sheet and find out if the switched VCC is present for the IC to run or not.

[DarkNeo]

Re: LG 42LE4500 - No 20V rail

IC101 is an ST L6599 Resonant Controller, datasheet here: https://cdn.badcaps-static.com/pdfs/...7cd0b455f4.pdf

[budm]

Re: LG 42LE4500 - No 20V rail

Look to see if you Voltage on the VCC pin against GND pin when PS-ON is present.

[DarkNeo]

Re: LG 42LE4500 - No 20V rail

Between the VCC pin (#12) and GND (#10) on IC101 I'm reading 13Vdc.

[budm]

Re: LG 42LE4500 - No 20V rail

Per spec sheet on page 8/36, the TurnOn threshold Voltage is 10.7V (typ.), so the IC may be bad or something is locking it out from running such as the Brownout protection such as pin 7, pin 8 on page 5/36. You need to test those Voltages as indicated on that page.

[DarkNeo]

Re: LG 42LE4500 - No 20V rail

Pin 7 is 1.87-1.9V and pin 8 is 0V. So it looks like the inputs to IC101 are good.

I also desoldered one of the transformers to check the coil arrangements and make sure there weren't any shorts/breaks. It all checked out and the bottom two pins on the secondary side are a single coil, so not seeing anything on the scope on these pins would indicate no/bad input to the primary coil.

So do you think it would be best to swap out IC101 and possibly the MOSFETs on the primary side?